Over the past few years, we have constructed large combinatorial chemical libraries [2] (>100,000 diverse and discrete small molecules) and carried out high throughput phenotypic screens of these libraries to identify small molecules that can control stem cell fate in several systems (introduced in Section B) [3]. Here we propose to develop and implement a high throughput screen of 100,000 diverse and discrete compounds to identify small molecules that can selectively induce differentiation of human embryonic stem cells (hESCs) [NIH registry code: WA01] into cardiac muscle precursor cells (which can further differentiate into beating cardiomyocytes). We will further confirm and characterize their effects and activities via various in-depth cellular/biochemical assays, and carry out structure-activity-relationship (SAR) studies of the selected hit compounds to optimize their potency and specificity. Collectively, the studies described in this proposal will provide novel chemical tools for producing human cardiac muscle precursors from hESCs for various applications, as well as studying the signaling pathways controlling cardiac specification, and may ultimately facilitate development of small molecule therapeutics to treat cardiovascular diseases, and/or stimulate cardiac regeneration in vivo. [unreadable] [unreadable] [unreadable]